Re: Rload

From: D. C. Sessions <dc.sessions@vlsi.com>
Date: Mon Aug 09 1999 - 17:48:00 PDT

"Montoya, Silvia" wrote:
>
> Hello IBIS modelers and Users.
>
> I am trying to determine what value to use for Rload. The default value
> is 50 ohms according to the IBIS spec.
> Why was this value choosen as the default value ?
> Is anyone using a different value and why ?

Rload needs to be small enough to show the driver reaching full current
(ie you don't learn much about the ramp up to 30 mA if you have a 1500
ohm Rload on a 3.3v part).

Rload also needs to be large enough to have a reasonable swing (1 ohm
Rload doesn't wiggle much when you hit it with 4 mA).

Because of the second-order effects (capacitive parasitic feedback from
the output to the predriver) the best accuracy will come from an Rload
near to the actual impedance seen by the driver. For a reflected-wave
driver this would be Z0. Another way to look at it is that the terminal
values for the typical VT table should be close to half of supply, which
keeps the min and max from violating the "too small" and "too large"
guidelines.

For most systems this comes to somewhere near 50 ohms, but don't obsess
about it.

-- 
D. C. Sessions
dc.sessions@vlsi.com
Received on Mon Aug 9 17:55:02 1999

This archive was generated by hypermail 2.1.8 : Fri Jun 03 2011 - 09:53:46 PDT